divagr18/SecureShell
Plug-and-play terminal security layer for LLM agents. Drop-in gatekeeper that prevents dangerous shell commands. Works with OpenAI, Claude, Gemini & more.
This project helps developers safely integrate Large Language Model (LLM) agents with command-line access. It acts as a "zero-trust" gatekeeper, evaluating every shell command an LLM agent tries to execute to prevent dangerous or inappropriate actions. The system takes potential commands from an LLM agent and outputs either the command's successful execution or a detailed reason for its blocking. It is for software developers and AI engineers building applications that give LLM agents the ability to interact with operating system commands.
Available on PyPI.
Use this if you are developing an LLM agent that needs to run shell commands and you want to ensure those commands are safe, platform-compatible, and well-reasoned.
Not ideal if you are an end-user without programming knowledge, as this is a developer tool requiring integration into an existing codebase.
Stars
22
Forks
3
Language
Python
License
MIT
Category
Last pushed
Jan 29, 2026
Commits (30d)
0
Dependencies
8
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/agents/divagr18/SecureShell"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
Windy3f3f3f3f/how-claude-code-works
Deep dive into Claude Code internals — architecture, agent loop, context engineering, and more....
Windy3f3f3f3f/claude-code-from-scratch
🔍 Claude Code 开源了 50 万行代码，读不动？用 ~3000 行 TypeScript 从零复现核心架构，11 章分步教程带你理解 coding agent 精髓。Build...
JCodesMore/ai-website-cloner-template
Clone any website with one command using AI coding agents
repowise-dev/claude-code-prompts
Independently authored prompt templates for AI coding agents — system prompts, tool prompts,...
Enderfga/openclaw-claude-code
OpenClaw plugin — turn Claude Code CLI into a programmable, headless coding engine with plenty...